Model-based reconstruction for simultaneous multislice and parallel imaging accelerated multishot diffusion tensor...

Area of Science:

• Magnetic Resonance Imaging
• Diffusion Tensor Imaging
• Image Reconstruction

Background:

• Multishot interleaved echo-planar imaging (iEPI) offers higher resolution for diffusion tensor imaging (DTI) compared to single-shot EPI.
• Prolonged acquisition time is a major limitation for iEPI in DTI applications.

Purpose of the Study:

• To develop and evaluate a novel model-based reconstruction technique for simultaneous multislice (SMS) and parallel imaging (PI) accelerated iEPI DTI.
• To significantly reduce the acquisition time of iEPI DTI while preserving image quality and accuracy.

Main Methods:

• DTI datasets were treated as 3D k-space data, undersampled in slice and phase encoding directions.
• Diffusion tensors were directly estimated via joint reconstruction of undersampled 3D k-space across all diffusion-encoding directions.
• A model-based formulation exploited correlations across diffusion directions to improve reconstruction.

Main Results:

• The proposed method demonstrated reduced estimation errors and artifacts compared to traditional parallel imaging reconstruction in simulations.
• In vivo experiments showed a fourfold acceleration (11 min 7 s to 3 min 53 s) for 4-shot iEPI.
• Image quality and precision of quantitative DTI parameters were comparable to fully sampled acquisitions.

Conclusions:

• The developed model-based reconstruction enables fourfold acceleration for iEPI DTI using SMS and PI.
• This method maintains high accuracy in tensor measurements, overcoming the time limitations of high-resolution iEPI DTI.